Critic's Notebook: Rethinking the Fascinating, Frustrating Trickery of Amazon's 'Homecoming'

Sam Esmail's directorial flourishes and the writers' tricky examination of memory were both brilliant and baffling, writes THR's chief TV critic.
Jessica Brooks

This column contains spoilers for the first season of Homecoming. Watch before reading.

Really good shows often make you think — and sometimes rethink. But there doesn't always have to be a negative connotation to that last bit. Perceptions not only shift when you're a critic; sometimes they take a while to solidify. 

All critics have found themselves giving a good review to a series based on early episodes — it's rare that we get a whole season in advance and even rarer that most of us have time to plow through a whole season before writing the review. Often we get two, three, maybe a handful of episodes. (Broadcast networks still routinely send only one episode, which is both archaic and unhelpful — but consider the source.) Periodically a series goes off the rails, creatively, in later episodes. All you can do is shrug.

I'm always happy just to finish a series. In the Peak TV era — and even with two critics (or more, in some places) shouldering the load — getting to the end of a series feels like a luxury and often happens by force, a driving need to find time to stop watching all those new series and their shiny beginnings and circle back to earlier shows to see how they all wrap up.

A lot of that is going on right now for critics, jamming in as much as they can before the end of 2018, trying to see (and finish) as many series as possible before making their best-of-the-year lists. 

I recently finished Amazon's Homecoming, from director Sam Esmail and creators and writers Eli Horowitz and Micah Bloomberg, who adapted the series from their popular podcast of the same name. The show stars Julia Roberts as Heidi, a kind of therapist hired by Colin (Bobby Cannavale), who works for Geist, a company we learn later primarily makes cleaning products. Colin and Heidi work at the Homecoming Transitional Support Center, a "transitional" place — but really an experiment — where returning veterans with PTSD issues are re-acclimated to society (but of course now that you've seen it you know it was to wipe the PTSD so the military could redeploy those pre-trained soldiers back to the battlefield).

What could go wrong? That is essentially the mystery.

Stephan James plays Walter, a soldier that Roberts begins to have feelings for, and there's a breakthrough moment when she realizes the drugs being tested on him have, indeed, wiped out his PTSD, but also most of his memory as well. It's the turning point in realizing the corporate malfeasance that has taken place and her role in it. 

It was a long journey to finishing the entire series — I got to review the first four episodes of Homecoming out of the Toronto International Film Festival, a full two months before the series even aired. In an almost unheard of (for me) rush, I ended up watching those first four episodes three times. As the review notes, I loved them. I was excited for the other six. But so much time passed before the season even started that I was off to a lot of other things. When I finally found the time to finish, it's not that my original opinion changed so much as entered a state of contemplative flux.

For starters, Homecoming is the perfect vehicle for Esmail, whose work on Mr. Robot redefined traditional television shot framing (so much so that the crew allegedly had to be re-trained for all the rule breaking) and who directs every episode of Homecoming, working intensely to bring a podcast with aural cues to life as a TV series with visual cues. He's basically trying to stay true to the concept while reinventing the structure (and there have been some plot tweaks as well). He pulled this off with the help of Tod Campbell, the Emmy-winning director of photography who worked on Mr. Robot with Esmail, and production designer Anastasia White, another Mr. Robot alum. In short, Homecoming was always going to be highly stylized.

Going into something like this requires being all-in on the Esmail aesthetic, which I mostly am. His visual conceit in Homecoming was jarring, as I describe here from the review: "There's a malevolent, Hitchcockian aspect to elements of Homecoming — something in the normal that's slightly off and menacing, just under the surface. This is particularly true when the series shifts to 2022 and Esmail employs a jarring aspect-ratio change, moving from 16:9 widescreen in the 2018 scenes to an old-school boxy 4:3 ratio in the 2022 scenes (and yes, flipping the trope seems intentional, just as making a time jump to 2022 seems almost humorous because, of course, nothing in the world has changed to some sci-fi model — it might be the least sexy but coldly original time jump in TV history)."

And: "At the same time, the bright hues and warm, upbeat music in the 2018 scenes give way to dull, depressing brown hues and staccato, threatening music in 2022. There's even a retro, flat feel to how Esmail and Campbell set up their close-ups in 2022, which almost mimic old TV series, while in 2018, similar to their groundbreaking work on Mr. Robot, characters are seen in modern, off-kilter close-ups or set into the lower corners of the frame. The result of both is riveting, despite calling attention to itself. Homecoming is a story that evolves in the writing, in what's spoken or unspoken, what's admitted and what's self-edited, and the robustly alive visual approach doesn't distract from that. The visuals and even the over-the-top musical choices enhance the creepy feeling of what went wrong at the Homecoming Transitional Support Center."

Ah, so here we are at the rethinking part. I worried about that jarring aspect-ratio change. But I went with it. Before I went all the way to the end, however, I had a long discussion (on stage at the Vancouver International Film Festival) with nine different writers and showrunners and we covered the idea of how the auteur TV director is now a real thing and how that might ultimately damage the actual writing element of small-screen storytelling. It was a conceptual discussion that I turned into a column (featuring their reactions to the auteur director surge) after finishing Netflix's Maniac and coming to the conclusion that that's precisely what happened with Cary Fukunaga's limited series, which was visually impressive but lacking in a compelling, believable story. 

The same fate does not befall Homecoming, but neither am I convinced (just yet) that it all works for me. Again, this goes to the idea at the top that some series make you think and then rethink, without that actually being a bad thing.

While Homecoming is a massive critical success (Metacritic has it with an impressive 83 rating, where only one of 35 reviews was even considered "mixed" while all the others were "positive" — including mine), I'm left with some concerns. Esmail's big reveal about the aspect ratio change comes in the eighth episode and it's a visually clever one. His explanation is that Roberts, whose character also had damaged memory, was in a "box," thus the aspect ratio change, and the conceit is dropped once she starts recovering her memory and we return to widescreen — with a visual panache worthy of Hitchcock. But that "reveal" is a lot of conceptual effort, especially since the message telegraphed to viewers without knowledge of Esmail's intent is that the aspect ratio changes signal a time shift only.

Beyond that, while plenty of people have noted that Esmail's use of sound included snippets from older movies (I'm not a huge Easter egg fan, but maybe that's because I don't seem to retain as many pop cultural reference points as others), I struggled there as well. As noted in my original review, I said the soundscapes were over-the-top, but since that was by choice, I went with it to see where it would end up. For me, it ended up being way too distracting. The sound took me out of the scene, repeatedly. 

Certainly part of that opinion came from hearing all those writers and showrunners talk about how any flourish that takes your mind off the show is not ideal. That said, I don't totally subscribe to that theory. I like directorial flourishes, provided that's not all there is (see: Maniac). If the story remains solid and does its job, and the direction is outstanding in whatever way it chooses, showy or subtle, that's a bonus. 

My concerns about the actual story in Homecoming have more wiggle room, because the overarching theme is memory — and memory is fallible. It's difficult to be critical of what the writers are playing at here because clinical evidence suggests memories run a very wide gamut from real to imagined, remembered and reconstructed, true to false.

But its Heidi's memory that is the most troubling from a story standpoint. We know that she purposely ate one medicine-laced meal with Walter (and since it was a "week six" meal, theoretically that pasta was pretty powerful). But Homecoming has us believe that the Heidi in 2022 is four years into a debilitating life of living at home with her mother (Sissy Spacek) and working as a waitress at a dingy waterside seafood restaurant, her memories hollowed out. From one meal? It's not implied that there were more meals — that one scene eating with Walter was more about the guilt she felt about her role in Geist's experiment so she was both self-harming and sabotaging the experiment. (Walter, pushed to eat more, becomes so incapacitated by the medicine that he can't be deployed.)

So if I'm willing to believe for dramatic purposes that Heidi's memory is coming back, which conveniently helps two big parts of the Homecoming plot, I can't also believe that one meal left her so damaged for four years that she couldn't remember who Colin was, much less identify him when he comes to see her (which of course sets up the ending of the first season as Heidi, apparently all better, goes in search of Walter on the West Coast and he ends up not recognizing her just like she didn't recognize Colin).

See, the fallibility of memory allows the writers to have Jeremy Allen White's character, Shrier, be so far gone, memory-wise, that he's the cautionary tale of the experiment, while Heidi in the end almost miraculously recovers hers and Walter is in some middle area where he's pleasantly clueless but also functional and at least pursuing a dream of peace he had long ago.

The twist at the end, where Walter talks to Heidi without recognizing her but tweaks her silverware just as he tweaked the pen on her desk back at the Homecoming Transitional Support Center, allows for interpretation. Was messing with the configuration of her silverware meant to send her a message that he really does recognize her? If so, why wouldn't he just say so (but of course if he did at that point, the very believable prior scene where he clearly doesn't recognize her would be a cheat — because nobody in the real world could be that good of an actor). Is it meant to indicate that on some level, without knowing it himself, something deep inside Walter recognized Heidi? That certainly works better as a story conceit. Again, the writers are playing with memory and it's hard to fault them on choices. As a positive example, I liked how Colin's wife revealed that the couple have a "box" where they write down something troubling that they did, that might lead to a fight or worse, and once it's put into the box and not talked about, the couple can move on. The parallel between the experiment and this couple's system of locking something away, or essentially destroying the memory and thus the reality of it, is an inventive touch (though of course nothing can excuse Colin's horrific preying on Heidi and sleeping with her).

I suppose contemplating the overall artistic success of Homecoming has something to do with where, precisely, I'm going to rank it on my best-of-2018 list. But I also think, constructively, it works as an example of how evaluating series can be fluid for different people. I'm not sure the aspect ratio change used to suggest Heidi's recovering memory was worth the effort, but I also thought it worked fine as a device to signal a time jump and wasn't ultimately that disruptive to the story (while the sound design was). The memory tricks? Still pondering that.